Every profession has its buzzwords to create the illusion that
things are more complex than they really are. Everything from the
used by medical doctors to the chatter of gearheads talking about
the latest car engine, simple concepts are often clothed in
professionals are no different in their use of complicated
nomenclature to describe simple things and ideas.
I know I was intimidated when I first heard the
. To me, it sounded like I would need a math Ph.D. or at least an
advanced understanding of statistical theory to figure out what
it meant. Not being an advanced math person, I was fortunate to
have had a trading mentor who patiently explained to me what
statistical arbitrage is and how to use it profitably.
Ever since I was made aware of this unique and profitable
trading technique, I have used it in a variety of
conditions to capture profits that would otherwise be
unavailable. This method's not for everyone, but if you're an
active investor who is looking for additional tricks of the
trade, statistical arbitrage may be just the ticket.
What Is Statistical Arbitrage?
, statistical arbitrage is a fancy term for pair trading, which
is the buying or selling of a pair of
based on their relationship with each other.
price of companies in the same sector or type of business follows
one another very closely. A pair trader observes the relationship
between two stocks and buys or sells whenever the relationship
gets out of sync, acting on the assumption that the historical
correlation is likely to continue.
Is it a foolproof method? No, but it does provide another
tactic in your investing toolbox.
It is easier to understand this concept with an illustration.
The following chart shows the relationship between
, perhaps the most popular stock pair for statistical
Notice how closely the two stocks follow each other until near
the end of May. At this time, Pepsico falls out of sync with
Coca-Cola, dropping as Coca-Cola stays steady and starts to
climb. Statistical arbitrage traders would purchase Pepsico stock
as soon as the divergence is recognized.
As you can see, the pair quickly moved back into sync,
opportunity for statistical arbitrage traders. There are
ways this can be approached.
For example, let's say Coca-Cola started rapidly climbing
higher than Pepsico. Savvy statistical arbitrage traders would
in anticipation of its price falling back into the historic
In addition, the idea is not just limited to two stocks. The
same idea can be applied to groups of three or more correlated
names. However, special software is often employed to manage
multiple-issue statistical arbitrage.
Here's another well-known pair trade:
As you can see, Target climbed out of the historical
correlation range on the chart. Traders invested in the pair
would short Target, holding until the historical correlation came
back into sync.
It's important to remember that it's not always obvious names
that present enough correlation to pair-trade. One example of
this is the relationship between
Other than trading on the same exchange, I cannot imagine why
two companies that are so diverse would be correlated so closely.
The reason could have something to do with the fact that
consumers may borrow
to buy Harley-Davidson motorcycles, but that's only a guess.
Risks to Consider:
Although closely correlated stock pairs generally come back
into sync with each other after diverging, there is no rule that
says this has to happen. Stock pairs can stay out of sync for a
substantial period of time, depending on the underlying
circumstances. Always use stops and position size properly.
Action to Take -->
Begin to chart the common pairs like Coca Cola and Pepsico,
General Motors (GM)
, and other closely related companies. In addition, experiment
with finding correlated pairs by simply charting a variety of
pairs of stocks. Although I like to use daily charts, tradable
correlations can be found in all timeframes. Professional traders
often use software, rather than visual charts, to find historical
pairs showing a statistical aberration from each other. Some
trading platforms have this ability built in, but this type of
software is readily available.
© Copyright 2001-2010 StreetAuthority, LLC. All Rights Reserved.